Like, share, vote

“A digitally empowered

civil society could

reconnect Europeans

with democracy…”

LIKE, SHARE, VOTE

Jamie Bartlett

Jonathan Birdwell

Louis Reynolds

Across Europe, low voter turnout in European and national elections is a growing concern. Many citizens are disengaged from the political process, threatening the health of our democracies. At the same time, the increasingly prominent role that social media plays in our lives and its function as a new digital public space offers new opportunities to re-engage non-voters.

This report explores the potential for social media to support efforts to get out the vote. It lays out which groups need to be the focus of voter mobilisation efforts, and makes the case for using social media campaigning as a core part of our voter mobilisation efforts. The research draws on a series of social media voter mobilisation workshops run by Demos with small third sector organisations in six target countries across Europe, as well as expert interviews, literature review and social media analysis.

Having affirmed the need for and utility of social media voter turnout efforts, Like, Share, Voteestablishes key principles and techniques for a successful social media campaign: how to listen to the digital discourse of your audience, how to use quizzes and interactive approaches, how to micro-target specific groups and how to coordinate offline events with online campaigns. This report concludes that, with more of our social and political lives taking place online than ever before, failing to use social media to reinvigorate our democracy would be a real missed opportunity.

Jamie Bartlett is Director of the Centre for the Analysis of Social Media at Demos. Jonathan Birdwell is Head of the Citizenship and Political Participation programme at Demos. Louis Reynolds is a Researcher in the Citizenship and Political Participation programme at Demos.

Like, Shar

e, V

ote

|

Jamie Bartlett · Jonathan Bir

dwell · Louis Reynolds

ISBN 978-1-909037-71-7£10

Supported by a grant from the

Open Society Foundations

Like Share Vote cover 3/11/14 7:03 pm Page 2

We produce original research, publish innovative thinkers and host thought-provoking events. We have spent 20 years at the centre of the policy debate, with an overarching mission to bring politics closer to people.

Demos is now exploring some of the most persistent frictions within modern politics, especially in those areas where there is a significant gap between the intuitions of the ordinary voter and political leaders. Can a liberal politics also be a popular politics? How can policy address widespread anxieties over social issues such as welfare, diversity and family life? How can a dynamic and open economy also produce good jobs, empower consumers and connect companies to the communities in which they operate?

Our worldview is reflected in the methods we employ: we recognise that the public often have insights that the experts do not. We pride

ourselves in working together with the people who are the focus of our research. Alongside

quantitative research, Demos pioneers new forms of deliberative work, from citizens’ juries and ethnography to social media analysis.

Demos is an independent, educational charity. In keeping with our mission, all our work is available to download for free under an open access licence and all our funders are listed in our yearly

First published in 2014 © Demos. Some rights reserved

Magdalen House, 136 Tooley Street, London, SE1 2TU, UK

ISBN 978 1 909037 71 7 Series design by modernactivity Typeset by Chat Noir Design, Charente

Set in Gotham Rounded and Baskerville 10

Cover paper: Flora Gardenia Text paper: Munken Premium White

Jamie Bartlett

Jonathan Birdwell

Louis Reynolds

Open access. Some rights reserved.

As the publisher of this work, Demos wants to encourage the circulation of our work as widely as possible while retaining the copyright. We therefore have an open access policy which enables anyone to access our content online without charge.

Anyone can download, save, perform or distribute this work in any format, including translation, without written permission. This is subject to the terms of the Demos licence found at the back of this publication. Its main conditions are: · Demos and the author(s) are credited

· This summary and the address www.demos.co.ukare displayed · The text is not altered and is used in full

· The work is not resold

· A copy of the work or link to its use online is sent to Demos You are welcome to ask for permission to use this work for purposes other than those covered by the licence. Demos gratefully acknowledges the work of Creative Commons in inspiring our approach to copyright. To find out more go to www.creativecommons.org

Acknowledgements 7

Executive summary

9

Introduction

15

1

Voter turnout trends in Europe: who

needs mobilising?

19

2

Social media and voter mobilisation: does

it work?

31

3

The capability gap: making the third sector

social media savvy

41

4

Principles and techniques of running a

social media voter campaign

49

Conclusion and recommendations

63

Annex: How did Twitter talk about the

European elections?

65

Notes 81

Acknowledgments

First and foremost we would like to thank Open Society Foundations for their generous support, without which this report would not have been possible. In particular we would like to thank Léonie van Tongeren, Peter Matjasˇicˇ, Heather Grabbe and Nadja Groot for all of their support and feedback during the research.

The project would not have been possible without the expert guidance of our Steering Group members. We would like to thank Mike Sani, Nick Anstead, Alberto Nardelli, Megan Bentall, Mevan Babakar, Erin Marie Saltman, Roman

Gerodimos, Philippe Marlière, Sarah de Lange, Christian Fuchs, Maria Sobolweska, and Dora Polgar for attending our Steering Group meetings and providing valuable feedback on early drafts of the report.

This report would not have been possible without the efforts of key people in the six European countries where we held workshops with third sector organisations. In particular, we would like to thank Peter Kreko, Oscar Moreda Laguna, Clotilde de Gastines, Adrian Chauvin, Sara Prestianni, Sano Lanucara, Paola Micalizzi, Ralph Du Long, and Hazel Nolan. We would also like to thank the many charities, social action groups and third sector organisations that hosted and attended our workshops across Europe.

Thanks and gratitude are due to Riccardo Liberatore, Rutger Birnie, Kyriakos Klosidis and Sofia Patel for producing excellent Twitter analysis reports for Italy, Netherlands, Greece and the United Kingdom, respectively.

As ever, this report would not be possible without the help and support of our excellent colleagues at Demos. Special thanks are due to Loraine Bussard for helping to organise the

workshops, and providing fantastic translation for the French workshop. Special thanks are also due to our colleagues at CASM, Carl Miller and Simon Wibberley at University of Sussex for their help coordinating the Twitter analyses in the case study countries. We also had fantastic support from a very talented group of research interns and assistants, including Peter

Fitzsimons, Kate Kipling, Emma Marshall, Richard Norrie and Chris Waller. Thank you all for your hard work. Thanks also to Alex Krasodomski-Jones for producing our ‘how to’ guide. Last but not least, we would like to thank Rob Macpherson, Daniela Puska and Ralph Scott for guiding the report through the publication process, and Sophie Duder and Alice Meaning for their help in organising the launch event.

The views expressed in this report represent those of the authors and not necessarily OSIFE. As ever, any omissions and errors are solely the authors’ own.

Jamie Bartlett Jonathan Birdwell Louis Reynolds November 2014

Executive summary

Electoral turnout is one of the most important indicators of democratic health. Low turnout undermines our democracy.1_In

national elections across the European Union, turnout declined from an average of 77 per cent in 1990 to 68 per cent in 2013.2

Turnout in European Parliament elections has fallen further, from 62 per cent in 1979 to only 43 per cent in 2014. While the entry of new states from Central and Eastern Europe accounts for some of this decline, it only partially explains it.

Part of the reason for declining voter turnout is that fewer people now trust political parties, join political parties or feel that it is their duty to vote. This is particularly true of younger generations. Yet, these trends do not necessarily suggest there is a loss of interest in politics or social issues by the public: they may instead be indicative of a shift towards new and different forms of political engagement, including online activity. Nevertheless, it is vital that efforts are made to reverse these long-term declines in voter turnout.

Social media may provide the answer. There are now almost 295 million social media users in Europe, comprising 40 per cent of the population. Facebook alone boasts 232 million active users. The average European spends around four hours online per day, much of the time on social media platforms.3

More than half of social media users use platforms like Facebook or Twitter to receive news and information.4

We have also seen in the past five years how social media can be used to engage citizens who feel increasingly estranged from the political system. The rise of populist groups in North America (eg the ‘Tea Party’) and Europe, including Beppe Grillo’s Five Star Movement and other political groups such as the Occupy movement, all demonstrate the political power of social media.

Mainstream political parties and civil society organisations, particularly in North America, have also used social media for campaigning for a number of years now. Civil society

organisations in Europe, on the other hand, still appear to lack the knowledge and skills to use social media to the fullest extent to influence their campaigns. This report is an effort to help address this gap. In particular, our aim is to assist third sector organisations across Europe working to increase voter turnout.

This report outlines:

Executive summary

· the latest trends in voter turnout in six European countries (chapter 1)

· recent evidence about social media’s ability to mobilise citizens for political engagement (chapter 2)

· the capacity of third sector organisations to use social media (chapter 3)

· key techniques and tools for designing and implementing a successful social media campaign (chapter 4)

· concluding comments and recommendations

We also provide a separate and detailed ‘how-to’ guide accompanying this report, which has been designed to give third sector organisations the basic skills they need to take advantage of social media for voter mobilisation.

Running an effective social media campaign: key findings

Below we outline four key principles and techniques to help civil society organisations develop effective social media campaigns for voter mobilisation: understanding audiences by listening to online conversations; using quizzes and interactive approaches; micro-targeting specific groups; and coordinating online campaigns with offline voter mobilisation events.

Listen to online conversations to understand your audience Social media not only allow activists to reach huge numbers of citizens, but they also provide an opportunity to hear the

perspectives of citizens as they debate politics and social issues. Gathering this information using new tools and techniques can help civil society organisations to design more effective campaigns. For example, Demos analysis of all the tweets containing European Union (EU)-related keywords across six European languages conducted for this report in the run-up to the European elections revealed a number of insights that could assist voter mobilisation efforts. For example, we found the following:

· Anti-EU populists dominated conversations on Twitter. · Policies are discussed, but primarily through the lens of

individual personalities.

· Humour and irony are used to frame political messages in unconventional language, which inspires positive sentiments and proactiveness.

While most social media analytic tools can be expensive, there are a number of cheap and free tools that third sector organisations can use to do this.

Use quizzes and interactive approaches to increase engagement

Tapping into existing uses of social media, through the use of online quizzes and interactive content, represents one of the greatest opportunities for activists to engage social media users. The largest study to date of social media and voter mobilisation found that interactivity and social influence must be at the heart of a successful campaign.

Across Europe, voter advice applications have demon-strated their ability to engage citizens who might otherwise not be attracted to politics. However, much more can be made of voter advice applications to maximise their use. In particular, we recommend developing a voter advice application that is designed for young people in style and content. If done correctly, such a tailored voter advice application could be a significant boost to turnout of youth voters because it taps into

two key behaviours of people on social media: love of quizzes, and doing something interactive.

Micro-target specific groups using social media advertising Through the exploitation of commercial social media advertising tools, it is possible to ‘micro-target’ specific groups with

campaign material tailored to the audience it is trying to reach. This ability dramatically increases the sophistication that voter mobilisation efforts can deploy. There is evidence that:

Executive summary

· the more personalised the message, the bigger the impact it has

· positive stories that are emotionally resonant are more likely to be shared

· appealing to how people identify themselves outside politics (eg as a student, through their favourite sport or sport team, through their support for a certain campaign), as well as the issues they are likely to care about, can help cut through the social media ‘noise’ and get their attention

Our research suggests that many third sector organisations still lack basic skills and knowledge around using things like social media advertising tools. In addition to the workshops we ran to provide organisations with these skills, we have also created an accompanying ‘how-to’ guide, designed so that even the most novice organisation can use social media advertising tools.

Coordinate online campaigns with offline voter mobilisation events Social media are a means to an end, not an end in themselves. Using social media effectively in conjunction with and in support of offline events is vital. There are a number of excellent examples of organisations that combine online campaigns with offline events in order to maximise impact, for example Bite the Ballot and Rock the Vote in the USA. As third sector

organisations become more savvy in their ability to run social media campaigns, they must remember that purely online campaigns risk succumbing to charges of ‘slactivism’, and have less impact overall.

Introduction

European and national political institutions are caught in a decline of trust and confidence that has steadily eroded their ability to inspire and mobilise voters.5_{Voter turnout in}

European Parliament elections has fallen year-on-year since 1979, from 62 per cent in 1979 to only 43 per cent in 2009 (although remaining stable for the 2014 election). Within these overall declines, there are also particular demographic groups that are even less likely to vote than average: young people (under 25s) and certain minority groups – including first or second generation immigrants.6

High electoral turnout is considered to be one of the most important indicators of the health of democratic systems.7_Low

voter turnout – and indeed the lack of political engagement that often underpins it –undermines our democracy and diminishes the legitimacy of elected leaders.8_{It can also impact – and skew}

– elections results.9_{Over the past decade an anti-establishment}

political insurgency has gained prominence and momentum in many European countries, most notably political parties like Front National in France, the Five Star Movement in Italy, Jobbik in Hungary, and UK Independence Party (UKIP) in the UK. These parties are often united in their critique of the EU and their hostility to immigration and minorities. While national polls (and elections) demonstrate that the majority of citizens in these respective countries disagree with the views of these parties, the lack of voter turnout in European elections, alongside their status as second tier elections in the eyes of the populus, allows anti-establishment populist political parties to do disproportionately well.

Because low voter turnout undermines the legitimacy of our democracy and our governments, increasing voter turnout has become a preoccupation for political parties, governments

and third sector organisations. While declining voter turnout is common across the developed world and is driven to a

significant degree by broad social trends, the variation between national voter turnout rates suggests that low voter turnout is a problem that can be solved by active intervention.10_{This report}

(and accompanying guide) examines opportunities to harness social media and new technologies to reverse the trend of political disenfranchisement, in order to provide the information that third sector organisations across Europe need to bring about effective voter turnouts.

According to recent estimates, three out of four Europeans use at least one social media platform and 60 per cent use one every day.11_{To put that into context, more Europeans access a}

social media platform daily than voted in the last European elections. Because of this extraordinary reach, social media are now a key part of political campaigns, affecting the way political parties form, organise, communicate and listen to potential voters.12

Political parties and data analytics companies have been using social media for insight into citizens’ concerns and voter mobilisation efforts for a number of years now, particularly in North America. However, these opportunities are not being exploited by civil society organisations, which often lack the technical know-how, time or resources to invest and use social media techniques. This report is an attempt to close this gap and to provide civil society organisations with the know-how and tools to use social media for their campaigns.

Method

In order to explore the potential for social media to mobilise voters, we have undertaken a comprehensive review of existing evidence. Throughout, we also drew on our own expertise, developed through the activities of the Centre for the Analysis of Social Media (CASM). In particular, we have pioneered the use of Facebook’s advertising tool to promote online surveys to supporters of new, emerging populist political parties and street-based groups across the political spectrum. In 2011, we published

The New Face of Digital Populism, based on over 13,000 responses from the activists and supporters of far right, anti-immigrant groups across Europe.13_{We have since continued these surveys,}

targeting supporters of far left groups and internet-based social movements. We have also developed software to analyse

conversations on Twitter about various political topics. Through our research efforts on social media, we have gathered a

significant amount of expertise about the most effective ways of targeting different groups of citizens effectively. We have engaged with a range of campaigners, activists and charities, and have conducted 13 interviews with social media experts and academics and reviewed over 120 papers, articles and reports.

We have also undertaken reviews about turnout rates and trends in six European countries: France, Greece, Hungary, Italy, the Netherlands and the UK. In each of these countries, we have drawn on available data to determine which groups – if any – consistently display disproportionately lower levels of voter turnout. This included data drawn from European elections, as well as national elections and other supplementary academic work. An expert steering group, whose members kindly con-tributed their time, feedback and recommendations throughout, also guided our research.

Based on these sources, we designed workshops that aimed to build the capacity of third sector organisations to use social media in an effective manner to accomplish their aims around voter mobilisation. Between January 2014 and June 2014, we conducted 12 workshops across Europe in six countries (France, Greece, Hungary, Italy, the Netherlands and the UK). In total we reached around 120 third sector organisations and campaigners.

Based on this research and practical training workshops, this report sets out:

· current trends in voter turnout across Europe, including key hard to reach groups (chapter 1)

· evidence on how social media can be used to understand voter concerns, communicate with disengaged citizens, and mobilise voters at low cost to increase voter turnout14_{(chapter 2)}

· current gaps in civil society capacity to use social media as a way of mobilising voters (chapter 3)

· specific techniques and (relatively cheap or free) tools that can be used to help better exploit social media, in order to improve voter turnout (chapter 4)

Social media mobilisation will become increasingly important as a method to mobilise voters in the coming elections – and for organisations to run campaigns of all types. It

potentially offers a vital way to identify people’s concerns, tailor messages according to their concerns and help nudge them towards the ballot box. With the right messages, platforms and know-how, this could be especially true for groups of citizens with historically low turnout. However, taking advantage of this opportunity is more difficult than it might first appear, and requires new skills, knowledge and, in some instances, software.

1

Voter turnout trends in

Europe: who needs

mobilising?

Across several measures, trust in formal politics in Europe has declined in recent years. Demos’ 2013 report Backslidersdescribed in detail the weakening of civil society in Europe, the decline in civic participation, and decreasing social and political capital.15

Part of this is due to the impact of the Eurozone crisis. Between 1999 and 2009, trust in EU institutions was relatively stable at around 45–50 per cent, but then dropped substantially from 48 per cent in summer 2009 down to 33 per cent in autumn 2012. There was also a notable increase in the percentage of Europeans who have a ‘negative’ image of the EU, rising from 15 per cent in 2006 to just under a third (29 per cent) in 2013.16

Politicians and political parties across Europe have experienced a wave of condemnation and distrust. In Italy and Greece, the percentage of the population who trust politicians currently stands at 15 per cent and 14 per cent respectively.17_In

the UK, the number of people who trust the government ‘at least most of the time’ dropped from 40 per cent in 1986 to 16 per cent in 2009.18_{In Germany, 64 per cent distrust political parties,}

compared with 89 per cent of French citizens and 79 per cent of British citizens.19_{While trust levels reached new lows following}

the 2008 recession and Eurozone crisis, data suggest there are longer term trends of increasing distrust towards traditional party politics and declining levels of engagement. According to calculations based on Eurobarometer data, trust in political parties in the EU has fallen by an average of 0.51 per cent a year since October 2003.20_{Similarly, average voter turnout for}

European Parliament elections has dropped by 2.7 percentage points per election since 1979.21

Political party affiliation is also on the decline. In 1980, 4 per cent of the UK population belonged to a political party,

whereas now formal party membership stands at around 1 per cent.22_{During the same time period, Italian parties lost around}

1.5 million members and French parties around 1 million.23_The

Netherlands experienced a decline of around 125,000 between 1980 and 2008, not insignificant given its population size, and Hungary has lost just over 40,000 since 1990. This pattern is not uniform, however; there has been an increase in party

membership in Greece of around 335,000.24_{These long-term}

trends suggest that declining political party membership is connected to broader shifts in social attitudes towards politics, the rise of issues-based politics and the advent of mass media.25

In the 2013 paper, ‘Virtually members’, Demos explored how social media was revolutionising political party membership and engagement.26_{Analysing social media followers for the three}

main political parties in the UK, we found that there were significantly more Twitter followers for the Conservatives (430,893) and for Labour (316,237) than formal party members (under 200,000 for each). We also found that there were

significantly more Facebook ‘like’s’ for the Conservatives and the Liberal Democrats than formal party members, but far fewer for Labour. Precisely how far virtual membership can replace traditional party membership remains to be seen. But as formal membership declines and virtual membership rises, it is clear that social media will become increasingly important for politics.

While levels of trust and party affiliation are not arguably essentialto the health of our democracies, the decline of voter turnout across Europe, at both national and European elections, cannot be ignored. In national elections across the EU, turnout gradually declined from an average of 77 per cent in 1990 to 68 per cent in 2013.27_{Voter turnout for the European Parliamentary}

elections – which has always been lower than national elections – has also continuously declined since the first election in 1979, when voter turnout stood at 62 per cent, down to 43.1 per cent in 2014.28_{It should be noted that part of this decline results from}

the reduction of the average by the entry of low turnout Eastern European countries into the EU, but that does not explain the continued – if less dramatic – decline in Western Europe. In the UK, voter turnout declined from 34.7 per cent in 2009 to 33.8

per cent, despite the significant attention the elections had in the UK press, driven by the rise of the UKIP and Nigel Farage.29

There are many reasons for lower participation in the Euro-pean Parliamentary elections than national elections, including the opacity of the EU’s day-to-day function, the ‘second-order’ nature of European Parliamentary elections, a perceived lack of satisfactory representation by parties involved in the elections, and growing cynicism with regards to the institutions of the EU.30_{With these issues in mind, it remains vital to the legitimacy}

of the EU, particularly in light of its growing legislative powers, that voter turnout does not fall too low.

Rather than reflecting a decline of interest in or engagement with politics, falling voter turnout could instead reflect changes in the manner in which people participate in politics (including through social media, which we discuss further in chapter 2). Nonetheless, reversing its decline must be an essential priority for government and NGOs in particular. Doing this in a strategic manner requires an overview of voter turnout across Europe and identifying priority areas. National laws and practices can lead to substantial variations in turnout. In Belgium, for example, where voting is compulsory (but where fines are generally unenforced), turnout in the 2014 European Parliamentary elections was 89.64 per cent.31_{In Slovakia, by}

contrast, turnout was 13.05 per cent, a record low in EU history.32

Perhaps even more important to the health of our democracy is the fact that, in some countries, there can exist a significant gap between the turnout rates of certain groups where our values dictate there should not be. These gaps exist between the rich and the poor, different ethnic minorities, the genders, and those living in different regions. For this report, we

conducted a comprehensive review and analysis of voter turnout trends across six different European countries and demographic groups. We discuss the key themes below.

Ethnic and religious groups

Academic research has highlighted the tendency for new immigrant groups and some ethnic minority groups to have

lower than average voter turnout.33_{This is often attributed to}

underlying characteristics – including, education, age or income – all of which increase the likelihood of voting among the majority population and are lower among minority populations in general. Cultural factors can also be at play, however, such as whether or not an individual was raised in the country, how many years they have lived in the country, their political knowledge and their social integration.34_{Studies of political}

participation among European immigrants show the influence of a lack of economic resources and lower integration.35

Determining whether these trends hold true in the context of European elections can be difficult to determine: data on participation in the European Parliamentary elections by ethnicity are not provided by the EU, and many countries do not keep national data on the voting patterns of ethnic minorities. However, there are some studies that help to provide an insight into the voting behaviour of ethnic or religious minorities.

In the UK, a study by Ethnic Politics suggested that, based on self-reported turnout, certain ethnic minority groups actually have higher voter turnout rates than British white citizens. While self-reported voter turnout is very often higher than actual turnout – in this case self-reported turnout was an average of 77 per cent, against a reality of 65.1 per cent – comparing self-reported voter turnout figures gives an indication of participation by ethnicity. Mixed race citizens were the least likely to report voting, with 65 per cent turnout. Conversely, those of Bangladeshi origin had the highest reported turnout (79 per cent), statistically equitable to the white population, who reported voting at a rate of 78 per cent. Overall, the study suggested there is a total variation in reported voter turnout of 14 percentage points between the ethnic groups with the highest and lowest turnout rates respectively (figure 1).36

In France, the situation appears quite different, though lack of ethnic and religious data restrict our ability to draw firm conclusions. One study highlights the variation between ethnic groups in voter turnout during the 2004 French regional elections. Using the 2003 Permanent Demographic Sample and the 2004 Survey of Electoral Participation, Maxwell found that

66 per cent and 70 per cent of people whose parents were born in France with French citizenship as a whole (something Maxwell called ‘Native French Metropolitan’) turned out in rounds 1 and 2 of the French regional elections respectively, while only 47 per cent and 54 per cent of citizens of Caribbean origin did. This suggests a variation of 17 and 18 percentage points between the highest and lowest turnout for different ethnic groups,38_{a significant variation in voting rates.}

It is important to recognise, as figures 1 and 2 show, that highlighting the total variation in turnout between the highest and lowest percentage can be misleading. While France and the UK have similar levels of variation (approximately 17 per cent and 14 per cent respectively), data from the 2004 French European elections suggest there is a clear discrepancy between the Native French Metropolitan population and ethnic groups from Northern Africa and the Caribbean.

Figure 1 _{Validated voter turnout in the UK by ethnicity in the 2010} general election

Source: Ethnic Politics, ‘Voting participation’37

Pe r ce n t ag e 60 65 70 80 Ethnicity 75 85 Bangladeshi White Black Caribbean Indian

All Ethnicities Pakistani

Mixed origin

The situation in Hungary is very different from that in the UK, France and other Western European countries, which have higher levels of ethnic minority populations and more diversity within their populations. In Hungary, the primary focus tends to be on the sizable Roma community rather than a range of different ethnic minority groups.

According to a 2011 EU Agency for Fundamental Rights report, self-reported turnout at the 2010 Hungarian national elections was 79 per cent for people who were not members of the Roma community, and 71 per cent for members of the Roma community, an 8-percentage point difference.40_{Other research}

has suggested that when controlled for socioeconomic status, education and other relevant factors, the Roma rate of turnout

Voter turnout trends in Europe: who needs mobilising?

Figure 2 _{Percentage voter turnout by ethnicity in France in the} 2004 European election

Source: Maxwell, ‘Political participation in France among non-European-origin migrants’39 Pe r ce n t ag e 50 35 40 25 20 45 30 Ethnicity Polish Portuguese

Native Metropolitan

Maghreb Caribbean

Italian

becomes largely the same as for other Hungarians.41_{By contrast,}

the same 2011 survey suggests there is a major disparity between the self-reported turnout of Roma and non-Roma respondents in Italy and France.42_{In the Netherlands, voter turnout is}

frequently lower among certain minority groups, a fact that is often explained as the product of lower average socioeconomic status and a lack of institutional trust.43

Region

Patterns in regional variation in voter turnout are similarly complex. In some countries, there is very little variation in voter turnout by region; in others there is a significant difference. This is often due to the manner in which regions within a country map on to socioeconomic status, education, and ethnic and religious variation.44

In the UK, for example, the further north one goes, the less likely citizens are to vote in the European elections (figure 3). During the 2009 European Parliamentary elections, there was notable regional variation: Scotland had the lowest turnout, with 29 per cent, while the South West turned out at 39 per cent. Similarly, there is a pronounced north–south divide in Italy. The Nord Occidental and Nord Orientale regions had turnout of 72 per cent, while the Insulare region, consisting principally of Sicily and Sardinia, had a turnout of 47 per cent.45_{In Greece, the}

smaller island collections Ionia Nisia, Voreio Aigaio,

Peloponnisos and Notio Aigaio varied from 43 per cent turnout to 49 per cent turnout, while Crete reported the highest voter turnout levels of all, at 59 per cent.46_{In Hungary, there was}

regional variation of up to 13 percentage points, between the 44 per cent achieved in the capital Budapest and the 30.9 per cent turnout of Jász-Nagykun-Szolnok.47_{On the other hand, there is}

significantly less regional variation in France and the Netherlands.48_{These regional differences should help voter}

mobilisation efforts to prioritise where their efforts should focus. Understanding possible regional variation is also important to ensure that improvements are measured from appropriate baselines.

Voter turnout trends in Europe: who needs mobilising?

Figure 3 _{Map of regional variation in voter turnout in the UK at the} 2014 European Parliamentary elections

South East 36.3% East 36% West Midlands 33.1% East Midlands 33.2% Yorkshire and the Humber 33.5% Scotland 33.4% Northern Ireland 51% Wales 32% North East 30.9% London 40.1% North West 33.3% South West 36.9%

Socioeconomic group

One of the most significant factors influencing voter turnout rates is socioeconomic status. Of our target countries, the UK has the most extreme relationship between socioeconomic status and voter turnout. According to the OECD Better Life Index, voter turnout in national elections for the 20 per cent who earn the most is around 73 per cent while for the bottom 20 per cent of earners, it is 50 per cent.49_{France has the lowest income}

inequality when it comes to voter turnout, with the top 20 per cent of earners voting at a rate of 89 per cent, and the bottom 20 per cent at 79 per cent.50_{The difference in voter turnout}

between the top and bottom 20 per cent of earners in Greece, Italy, Hungary and the Netherlands is 12, 12, 19 and 19 percentage points respectively.51_{As noted above, variations in}

voter turnout that appear to correlate with ethnicity and regional location may in fact be driven instead by differences in

socioeconomic status.

Age

Another highly visible and often discussed demographic disparity when it comes to voter turnout is age. On average, turnout in the EU is lowest among those aged 22–25, at 32 per cent, and highest among those aged 70–73, at 48 per cent.52_This

age disparity is particularly pronounced in the UK. According to the Electoral Commission’s report on the administration of the 2009 European Parliamentary elections, 74 per cent of people aged 65 or over reported voting, compared with just 13 per cent of 18–24-year-olds.53_{Moreover, this is not just what sociologists}

refer to as a ‘lifecycle effect’ (something which characterises ‘young people’ of every generation), but rather there appears to be a ‘cohort’ effect at work in that the ‘youth’ vote appears to be declining from each generation to the next; the percentage of young people turning out to vote in the UK has declined significantly since the early 1990s.54_{General election turnout}

among 18–24-year-olds in the UK has fallen from 75.4 per cent in 1992 to 44.3 per cent in 2005, falling 5.1 percentage points from 2001 to 2005, while electoral participation increased among all other age groups.55

Even this relatively uniform trend varies from country to country. For example, although during the 2009 European elections in France, Greece, Hungary, Italy and the Netherlands 18–24-year-olds were the least likely to vote, in the UK it was 25–39-year-olds who were the least likely to. In the UK, the largest gap between age groups was 32 per cent. In Italy, it was only 7 per cent.

Gender

A 2009 European Parliament survey found that men and women were equally likely to have participated in the previous national elections in respective nations, and men and women were also equally likely to have voted in the 2009 European Parliamentary elections.57_{However, our review suggests that there are}

differences in turnout along gender lines in some countries. Figure 5 shows the breakdown of voter turnout differences in

Voter turnout trends in Europe: who needs mobilising?

Figure 4 _{Voter turnout by age in the 2009 European elections}

Source: European Parliament, EB71.3 European Elections 200956

Pe r ce n t ag e 50 60 70 40 30 10 0 20 18–24 25–39 40–54 55+ Hungary Netherlands France UK Italy Greece

European elections across the six countries we analysed. For example, in Greece men commonly outvote women by around 10 percentage points in national elections. Thus, organisations working on voter turnout in Greece may want to prioritise their efforts on women. In Hungary by contrast, women outvote men by 1 per cent.58

Summary

Increasing voter participation in aggregate is not the only goal that we should aspire to. Just as important is ensuring that the electorate is representative of the population at large and does not systematically under-represent certain groups.60_{In many}

countries across Europe, to varying degrees, a number of minority and marginalised groups are less likely to vote. If a democratic institution seeks to represent a multicultural society, Figure 5 _{Differences in voter turnout rates in 2009 European}

elections based on gender

Source: European Parliament, EB71.3 European Elections 200959

Pe r ce n t ag e 50 60 70 40 30 10 0 20 Male Female

or even the needs of a society as a whole, it should pursue an inclusive electorate. Prominent voter turnout variations between different groups to differing extents warp the shape of the electorate and ultimately reduce the representative value of a government, as well as our collective confidence in it. In the next chapter we look at the possibility social media hold for

redressing these imbalances in voter turnout.

2

Social media and voter

mobilisation: does

it work?

Every new technology – from polling, to radio, television and email – has been picked up by political strategists and campaigners to reach voters and mobilise them to vote. In the 2004, 2008 and 2012 US presidential elections, mass mailings became virtual as political campaigns made extensive use of emails to communicate with supporters and potential voters. In 2008 and 2012, Facebook, Twitter and other social media sites became electoral battlegrounds. In addition to providing a direct and potentially cheap way to communicate with voters directly, social media sites are valuable because they are key arenas where citizens have discussions about political parties, candidates and specific issues. These conversations can provide a source of valuable information for political campaigns if they can be tapped into. While many campaigners continue to stress the importance of face-to-face communication, new technologies and social media are now vital to campaigns – including the

facilitation of face-to-face engagement.

In the 2000 US presidential elections, it was demonstrated that elections could be won and lost with very small percentage changes of certain key demographics or voters.61_{In 2004, the}

Republican Party introduced the micro-targeting techniques that had been developed in market research into politics. Political micro-targeting involves creating large databases of information on individual voters that combines voting histories, contact information, and demographic and consumer information to form complex profiles. These data are then analysed through statistical modelling in order to understand the political preferences and attitudes of individual voters, and to prioritise resources towards persuadable voters and facilitate the personalisation of messaging delivered to them.62

Micro-targeting allowed the Republicans to make better use of their resources, avoiding those voters unlikely to support the Republican ticket and instead deliver personalised messages to likely voters.63_{By the 2008 election, the Democrats}

had followed suit and surpassed the micro-targeting efforts of the Republicans.

The 2008 and 2012 US elections are considered watershed events in the use of digital technologies for micro-targeting. The Obama campaign ran a so-called ‘cave’ headquarters in Chicago comprising teams of data scientists who fused conventional polling with online and offline data to produce a series of predictions for how every American voter would act on Election Day. (Quite how Obama’s campaign succeeded in this dramatic fertilisation of politics with big data was a closely guarded secret.)64_{Although micro-targeting has been an important}

electoral technique since the early 2000s, it became increasingly important in the 2008 and 2012 US elections. In the 2008 presidential election, Republicans and Democrats spent a combined $20 million dollars on social media. In the 2012 election, that sum had risen to $78 million.

Although solid evidence of social media strategies in politics outside the USA is fairly limited, there are signs that certain European parties are looking to follow the American example. In 2013, for instance, Labour hired Matthew McGregor, Obama’s ‘digital attack dog’ in the 2012 election. McGregor was known for producing slick videos exploiting gaffes made by Republicans – often within minutes or hours of them being reported. These videos were published and shared on Facebook and Twitter and reached millions of Americans. The UK Conservatives are revamping a big database called ‘Merlin’, while Labour have ‘Nation Builder’ and Blue State Digital Tools, and the Liberal Democrats have VAN (‘Voter Activation

Network’). The data comprising these databases are gathered from a variety of sources, including the electoral register, local party activists and public and commercial sources.65

Box 1 _{Obama 2012 and the network effect}

In the final weeks of Obama’s 2012 re-election campaign, Obama’s digital team realised that approximately half of their targeted swing-state voters under the age of 29 had no listed phone number. So in response, Obama’s digital team developed a Facebook app that more than 1 million people would eventually sign up for. These users gave the Obama campaign permission to look at their Facebook friend lists, granting them access to previously unseen voters. Around 85 per cent of people without a registered phone number could be found on these Facebook friend lists. Obama’s team called this ‘targeted sharing’, and in the final weeks of the campaign they bom-barded subscribers with requests to share online content at the click of a mouse. Over 600,000 users did so, reaching over 5 million contacts. People were asked to vote or register to vote, donate or simply watch an Obama campaign video. The nature of the appeals changed according to the potential voter’s profile.

According to Timemagazine, initial tests of this strategy found that it resulted in statistically significant changes in voter behaviour.66_{People whose friends sent them requests to}

vote, for example, were more likely to do so. This confirmed the results of earlier studies: that people tend to respond better to messages from friends and contacts than to official bodies and campaign groups.

Social media are increasingly vital to campaigns, political or otherwise, because they allow campaign managers to better understand voter concerns and preferences, to circulate campaign content and messages, and thus to better reach and mobilise voters. Recently, we have seen how social media can be used to engage citizens who feel increasingly estranged from the political system, through a more organic, less hierarchical social media model more characteristic of social movements. This type of social media exploitation has played a critical role in the rise of populist groups in North America (eg the ‘Tea

Party’) and Europe, including the likes of Beppe Grillo’s Five Star Movement.

Social media and issues-based activism

Social media may also be contributing to and shaping changes to political habits of many European citizens. In particular, as noted above, voters and voters alike are increasingly non-partisan: less likely to be bound by tightly defined political ideology or even a defined identification with one party or another. With the internet providing more information on a range of different issues, citizens may feel more empowered to come to their own views about various issues, rather than simply subscribing to a political party’s manifesto.

Moreover, social media have diversified the spectrum of political engagement, allowing for engagement that could best be described as ‘micro-activism’: small-scale, issues-led, many-to-many political communication.67_{According to Jose Marichal,}

users of social media platforms can now use the technology to explore and define their political identities in a more nuanced manner, outside traditional political frameworks. Through commenting on news stories, signing petitions, engaging in debates, re-sharing opinions and liking groups or pages, social media users both define their own political identity and publicly display it to their peers.68_{According to Marichal, social}

networking sites can even ‘help encourage the formation of political identities that may force users to reflect more deeply on themselves as civic beings’.69

While online political action is sometimes derided as ‘slacktivism’, research shows that online activity can stimulate offline political activity and collective action.70_Evidence

suggests that a high level of political engagement among young people online is correlated with engagement in offline activities.71

In this project we wanted to explore how well suited social media was at involving those at the margins of the voting process: the young, recent immigrants and ethnic minorities, many of whom tend to be active users of social media.

Social media topography and use among

target groups

With every passing year, social media become more engrained into our lives. There are almost 295 million social media users in Europe, 40 per cent of Europe’s population. Facebook alone boasts 232 million active users. The average European now spends around four hours online per day, much of the time on social media platforms.72_{In the Netherlands the average time}

spent on social media per day is just over an hour and a quarter. In the UK and France, it is over one and a half hours a day, and in Italy, two hours.73

The ubiquity of the internet and social media is changing the way people get their news. With newspapers migrating online, over three out of four British internet users turn to the internet in order to access news information. Even more import-antly, more than half of social media users use social media sites like Facebook or Twitter to receive news and information.74

European studies on demographics and political engagement online have shown that young people are more strongly represented on social media than other groups, and that young people respond particularly well to political engagement on social media.75

There is some variation by gender across platforms, but women use social media at roughly the same volume as men. Statistics vary, but the Oxford Internet Institute’s 2013 figures suggested that women use Facebook and YouTube very slightly less than men, but Twitter significantly more.76

It is sometimes argued that access to the internet – the divide between the ‘haves’ and ‘have-nots’ – could exacerbate the democratic deficit that affects people from lower socioeconomic groups. If the political world gravitates to the internet, those without internet access can be cut off from politics. Yet recent research published in the Social Science Computer Reviewhas suggested that greater levels of internet access and engagement online increases political participation much more effectively for members of lower income groups than it does for those in higher income groups.77_{This study highlighted correlations between}

high levels of internet access, socioeconomic status and levels of political engagement. However, the study also found that groups

from lower socioeconomic backgrounds experienced the largest increase in political involvement and knowledge the more their access to internet increases.78_{In other words, despite access}

limitations, it appears that the internet may be helping to close the gap that has always existed between social class and political involvement.

The new ‘big data’ analytics

There are a growing number of academic and commercial efforts to make sense of social media data sets for research or (more typically) advertising and marketing purposes through big data analytics – the analysis of data sets so large and complex that they are difficult to process and analyse through traditional means.79_{Increasingly, big data analytics are a critical element of}

election campaigns. This involves the collection of large data sets about citizens’ views and behaviour to give politicians a more detailed picture of voters’ attitudes, beliefs and behaviour. According to the UK National Endowment for Science,

Technology and the Arts (Nesta), ‘what had once been done with pen and pencil is now being done in real time and at a staggering pace thanks to innovative data technologies’.80_{Although some of}

it is quite sophisticated (and expensive), there are a growing number of free or cheap tools that can be useful for gaining insight into voters’ concerns and attitudes.

Micro-targeting and personalisation

The most important aspect of social media campaigning is its ability to conduct ‘micro-targeting’, and in doing so personalise content. The Yale Institute for Social and Policy Studies

observed that ‘across a number of mobilisation experiments, one consistent finding is that more personalised messages are more effective in mobilising voters’.81_{The more closely crafted to an}

individual identity the message is, the more likely that people who self-identify with that group will pay attention to the campaign. A generic message, on the other hand, is less likely to be clicked on.

Social media are extremely useful for micro-targeting efforts. First, they allow advertisers to reach people directly based on particular demographic and belief characteristics. Users of social media often willingly provide demographic information, such as age, gender, education and marital status. Geo-location data can also pinpoint a user’s location. Moreover, users often ‘like’ various products, movies, books, political parties and other campaigns. Conversations and comments can be searched for, analysed and traced back to specific social media users. In other words, there is more information than ever before at the fingertips of advertisers – or election strategists – to understand different audiences. This regular personalisation of messaging is ethically sensitive, raising important questions relating to informed consent. Advertising tools can also allow groups to be targeted with adverts based on this information – with the potential for highly personalised messages.

Second, recent research into voter turnout has found that social pressure is a significant motivating force on citizen behaviour; for example, if you believe your neighbours or friends voted, then you are more likely to.82_{Social media provide}

a useful vehicle for such social pressure, because they allow people to show others that they have voted, thereby increasing the sense of social duty that others feel about voting. A recent example of the exploitation of this social pressure – as well as of social network-facilitated information distribution – played a role in the Scottish referendum. Facebook, which a few days before the vote claimed to have had over 10 million comments, likes and posts related to it on their platform, introduced an interactive ‘I’m a voter’ button for the Scottish referendum.83

Appearing at the head of Scottish users’ news feeds on

referendum day, it gave the option for users to either access more information about the vote, or share with their Facebook friends that they had voted. Perhaps more innovatively, the month before the vote Facebook launched a ‘registered to vote’ live event to encourage users to register in time for the referendum.84

Applying this insight to the task of voter mobilisation, it could be argued that the key is mobilising a few people within networks that tend to have low voter turnout. If a sufficient

number of people can be convinced to vote and broadcast the fact that they are voting through their networks, it could encourage others in their networks to get involved as well.

However, social media are not only useful for micro-targeting voters directly. Social media platforms and analysis tools also offer ways to create community groups, run campaigns or reach out to people at low or zero cost. For example, in 2000, Rock the Vote’s voter registration application, an online tool to facilitate voter registration, registered almost 165,000 new voters. In 2004, 1.2 million voter registration forms were downloaded from their site, and by 2008 it was 2.5 million.85_{In February}

2014, Bite the Ballot launched the UK’s first ever National Voter Registration Day (NVRD), in a digital campaign that stretched across YouTube, Google+, Facebook and Twitter. Bite the Ballot’s campaign led to the registration of around 35,000 young people.86

Making the most of social media’s potential

The political potential of social media in Europe is not sufficiently exploited. According to Alberto Nardelli, who runs technology and news company Tweetminster and the Guardian’s Data Blog, ‘social media has the potential to mobilise people in terms of turnout and that potential is linked to the context of a country; it is a potential that is not always realised’.87_Similarly,

Mike Sani, head of the voter mobilisation group Bite the Ballot affirms this potential, and argues that ‘our democracy needs to evolve at the pace that our technology and more importantly our communication has evolved’.

Although social media are becoming a huge part of political life across Europe, there are not many examples of it being used in any meaningful way for voter mobilisation efforts. In probably the most notable example, Beppe Grillo used social media to recruit and organise ‘meet-up groups’, which led to his new party going from its founding in 2009 to winning a quarter of seats in the Italian Parliament at the last election. Meet-up groups are free online chat groups that can be set up by anyone to coordinate offline activities. Using social media to set up

meet-up groups, Grillo was able to create a large network of supporters across the country, who were then committed to campaigning for him during the election. In total, Grillo had over 1,400 meet-up groups worldwide, comprising at least 170,000 people.88

One of the more sophisticated efforts at engaging young voters has been the Dutch Verkiezings app, which is deployed on Facebook, Twitter and Instagram. The app displays at a glance where the major parties stand on issues that the user cares about, and provides news and information related to the election. Social media users receive ‘push notifications’ as the election

approaches with details of the precise timings of the election. The application also facilitates issues-based political meet-ups. The app is designed to be run by municipal councils, regional and national governmental bodies, and provides detailed feedback regarding page views, young people reached and election reminders distributed. A September 2012 pilot scheme in the municipality of Doetinchem suggested some success, notwithstanding some notable difficulties demonstrating an actual effect and a causal relationship between the social media campaign and increases in voter turnout.89

Compared with the USA, the use of social media by third sector organisations and governmental institutions in Europe is still small scale. According to Nick Anstead, co-author of ‘The 2008 digital campaign in the USA: the real lessons for British parties’,90_{‘social media changes the game’ but ‘there is room for}

extensive innovation in European politics’, particularly in light of the failure of institutions to adapt to social media and the political practices of twenty-first-century citizens.91

Summary

Social media offer a potentially valuable and useful new way to mobilise voters. While not a panacea to decreasing levels of political engagement, if used carefully and thoughtfully, they could provide a cheap and effective supplement to campaigns that aim to encourage political engagement. In particular, there are a number of free and easy to use tools that can be employed

by third sector organisations to improve the way they reach out to and connect with voters. These tools – and how to use them – are explained in the ‘how-to’ guide that accompanies this report. In the next chapter we outline the workshops that we conducted with NGOs and activists in six European countries. These workshops provide a window into capacity levels of NGOs in Europe trying to grapple with the challenge of making effective use of social media and the internet.

3

The capability gap:

making the third sector

media savvy

In March and April 2014, Demos delivered 12 workshops in France, Greece, Hungary, Italy, the Netherlands and the UK (as well as a workshop in Barcelona, Spain) on voter turnout and social media. In total, we presented to over 120 NGOs, campaigners and activists who are working on voter mobilisation, countering xenophobia and racism, and

supporting migrant and marginalised groups across Europe. The majority of the organisations who attended our workshops are conducting campaigns following our training or are extending ongoing campaigns.

In some instances, the organisation representatives who attended the workshop specialised in social media, while others had only a passing knowledge or familiarity. The workshops were designed to be useful and informative for all levels of social media expertise. You do not need to be an IT expert or social media aficionado to make effective use of the tools that social media have to offer. Too often older activists dismiss social media as key to their efforts through bias or a perceived lack of

knowledge. Our ‘how-to’ guide is written with these social media sceptics and novices in mind.

W

orkshop design

The workshops were designed to help these organisations improve their ability to mobilise voters and run campaigns using social media. However, the workshops also provided an

opportunity for the Demos research team to investigate and explore the current capabilities of third sector organisations to undertake social media campaigns and what sort of assistance they require. The results of these investigations informed the focus of the accompanying guide.

The workshops were designed to include presentations from Demos researchers, as well as practical exercises for

participants to make use of social media advertising in particular. The content of the workshops included:

· the overall electoral landscape in each country, including trends in voter turnout, mobilisation levels and specific problem areas (presented in chapter 1)

· the potential uses of social media, including national penetration levels and targeted advertising tools that are available at low cost (presented in chapter 2)

· CASM’s analysis of political and social discourse on Twitter, providing an insight into the concerns and priorities of the public and a means by which to inform messaging and campaigning efforts (available in the annex to this report) · a detailed overview of how to use Facebook and YouTube

advertising, DIY social media toolkits and voter advice applications (presented in chapter 4 and the ‘how-to’ guide)

In each workshop we divided participants into three or four groups and gave them a practical exercise: we assigned each group a theoretical population to target with the tools we provided (eg young men from ethnic minority backgrounds living in large cities). We provided a series of questions for them to consider while targeting their assigned group, for example: · Which tool or social media platform is the best to reach your

groups?

· What regional areas should you target? · What interests would you target?

· What are the interests, key words and demographics that seem most suited to reach your audience?

· Which images would you use? · What would your message(s) be?

· How much does it cost – and how does this compare with alternative methods?

Using Facebook to target voters using geography,

imagery and interests

Facebook allows users to run campaigns that target users on the basis of demographic and interest-based subjects. Thus, a key component of the workshops involved teaching participants how to use Facebook’s advertising tool to find hard to reach citizens who traditionally do not vote.

In order to choose which geographical areas to focus on, workshop groups had to conduct desk-based research using various sources, including their country’s census data and the Facebook advertising tool. For example, election studies could be used to determine which areas had the lowest levels of voter turnout, or a preponderance of a certain target group;

Facebook’s advertising tool can be used to identify which cities contain the largest number of young men who say they are interested in Islam.

When it came to selecting images that might appeal to these target groups, the groups were able to draw on the insights from the research we presented (eg the importance of human faces, and appealing to specific identities) and their own experience, and to be more creative in selecting a range of different images which could be A/B tested (this is called ‘split testing’) for effectiveness. In the Netherlands, for example, the workshop group who were targeting young, Dutch Muslim men chose images that ranged from a picture of intertwined Dutch and Moroccan flags, to a picture of the European flag, a picture of a famous Dutch Moroccan football player, and a picture of a young Dutch Moroccan holding his Dutch passport. As we explain in more detail in the accompanying guide, Facebook’s advertising tool then measures which image receives the most clicks, allowing the user to further refine the effectiveness of their advert by adding similar versions of the most effective images.

Choosing the interests to target allowed the participants to see the benefits of the Facebook advertising tool. For example, once the first group identified their demographic targets, they selected those within the demographic group who were interested in ‘Islam’ or the ‘Quran’. The Facebook advertising tool then immediately suggested a number of additional interests that were correlated with those characteristics already selected.

In this instance, the top correlated interest was ‘boxing’. Armed with this insight, the organisation running the advert could ensure that some of the images used in the targeted advert included images of boxers (specifically Dutch Moroccan

boxers). Moreover, it could also suggest that a voter mobilisation campaign targeting this group should seek to complement their online efforts with events or posters placed around boxing gyms in cities like Amsterdam, Rotterdam and Utrecht.

W

hat are third sector organisations doing at

the moment?

The workshops allowed us to get a sense of just how savvy small- to medium-sized third sector organisations are at using social media to support their work. Many organisations exhibit significant technical competence in social media campaigning and organisational focus on it, while others are far less active on social media. Nevertheless, across the third sector organisations in our six target countries, broad similarities and trends were apparent.

Almost all the participants in our workshops had a presence on Facebook, Twitter or both, with a wide variation in levels of activity and interactivity on these sites. It appeared that use of social media was an incidental activity as opposed to a part of the wider strategy of many organisations. A significant minority of organisations did not actively use social media. A few had no substantive online presence. Some were pursuing (or had previously pursued) active social media campaigns involving relatively sophisticated digital tools, like online quizzes or voter advice applications.

Very few organisations and participants had ever used an advertising tool for any purpose, or used any more sophisticated tool or application. Almost none had used any type of social media analysis to understand their target audience.

Despite the relatively low level of this activity, activists almost universally acknowledged the potential utility of social media efforts, and often expressed a strong desire to ‘upskill’ in that regard. When taught the core dynamics of social media

advertising, the key principles of a successful campaign and the basic technical skills required to undertake social media

advertising, activists were keen to advance their own social media campaigns, and often discussed at length their plans for

innovative online efforts.

It appeared that many representatives from the organisa-tions we worked with appreciated the usefulness of social media advertising but needed a significant amount of time for practical exercises in order to derive concrete insights into its use for their own campaigns. This suggests that embedding best practice in using social media would require an extended period of practical exercises. Running a social media advertising campaign that is effectively targeted is a resource-intensive commitment, even though it can also be highly cost-effective. In order for organisations to conclude that it would be in their interest to invest the needed resources, further sustained work and support is needed to help them develop a familiarity with the tools and get a sense of their effectiveness.

Some campaigners expressed concerns regarding their ability to undertake social media campaigning in a more significant manner. This was predominantly for two reasons. First, though social media advertising is a relatively inexpensive means of promoting a campaign, it still costs something, and a number of campaigners had no budget at all, often campaigning in their spare time and without payment. Second, some were worried they would struggle to find time for social media campaigning. It demands brief yet regular periods of time for management, and though this time demand can be reduced through free management tools, it was a commitment that many could not meet.

W

here do third sector organisations need

further support?

We identified four key areas where organisations needed training and support – using cost-effective cheap or free tools, measuring success, making the most of events and having two-way

Using cost-effective and cheap or free tools

The most common type of feedback from participants related to the availability of free tools. Many participants complained that, although they knew of excellent and sophisticated techniques to collect or analyse social media data sets, or run campaigns, these tended to be too expensive. There was a feeling among

participants that there were free or cheap tools available that could be useful, but no single resource where third sector organisations can find them, and learn how to use them

effectively. Although we demonstrated to workshop participants that effectively designed adverts (those that receive a lot of clicks) could be extremely cost-effective, managing these campaigns can be resource intensive, and may require too much time management for small NGOs.

Measuring success

The second most frequently asked question we received related to measuring success beyond ‘likes’ and ‘retweets’ Indeed, most companies, organisations and even politicians are asking this question in this new era of social media. How does clicking ‘like’ – or ‘retweeting’ something – relate to purchasing a product, taking the extra step to demonstrate for a cause or even voting in elections? Almost all of the organisations solely measured the success of their social media campaigns on the basis of ‘likes’, ‘shares’ or ‘retweets’. There was little to no understanding of how these social media measures translated to success in the ‘real’, offline world. T